Skip to content
View MARUCIE's full-sized avatar
1 follower · 2 following

Achievements

Achievement: Pull Shark

Achievements

Achievement: Pull Shark

Highlights

  • Pro

Block or report MARUCIE

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
MARUCIE/README.md

Maurice Wen

AI Agent Systems Engineer & Data Intelligence Architect
Building reliable agent infrastructure for enterprise workflows

Email

About

Senior product and engineering leader focused on agent runtimes, multi-model orchestration, context systems, and domain-specific AI products.

Core thesis: AI agents need infrastructure, not just prompts. The durable layer is the runtime, control plane, evaluation system, knowledge boundary, and operational evidence that make model capabilities reliable.

Private Platform Infrastructure

System Role
Armature Agent SDK Canonical private agent runtime and SDK foundation: provider-neutral execution, tools, MCP, workflows, memory, evaluation, and multi-model collaboration
AI-Fleet Private control plane for heterogeneous coding agents, skills, routing, evidence gates, and operational automation
Cognebula Enterprise Private enterprise context brain built around Graph RAG, knowledge operations, and deployable infrastructure

Armature owns reusable agent execution capability. AI-Fleet owns orchestration and completion governance. Product repositories consume these foundations without creating parallel control planes.

Selected Public Work

Project What it does
AuthBox Identity and credential infrastructure with an AI-agent gateway
Babel Bench Classical-language LLM safety benchmark, evaluator, and defense catalog
OpenClaw Foundry Curated AI-agent skill marketplace with evidence-based ratings
Claude HTML Style Router Routes HTML deliverables to genre-specific design systems with CJK-aware typography
SOTA Agent Hub Searchable directory of agent skills, MCP servers, prompts, and tools
Lingque Ledger Production delivery surface for the Lingque finance and tax WorkBuddy

Focus Areas

Agent Infrastructure   Agent runtimes · MCP · Multi-model routing · Skills · Evaluation
Knowledge Systems      Graph RAG · Context engineering · Evidence-bound memory
Product Engineering    TypeScript · Python · Swift · Cloudflare · Docker
Enterprise Domains     Finance and tax · Compliance · Data platforms · Workflow automation

Background

  • Enterprise data-platform and AI-product leadership
  • National-scale operations and monitoring systems
  • Privacy-preserving data collaboration and federated learning
  • Intelligent finance, tax, compliance, and invoice-processing systems

Maurice | maurice_wen@proton.me

Popular repositories Loading

  1. authbox authbox Public

    Your keys. Your identity. Unstoppable. -- Seed-phrase password manager with AI agent gateway.

    HTML 5

  2. babel-bench babel-bench Public

    Ancient languages break every major LLM safety filter. Benchmark dataset + evaluator + defense catalog. Based on CC-BOS (ICLR 2026).

    Python 3 1

  3. cloud-claw cloud-claw Public

    Cloud-Claw OpenClaw Gateway - Cloudflare Workers Containers

    TypeScript 1

  4. openclaw-foundry openclaw-foundry Public

    The curated AI Agent skill marketplace — 37K+ vetted skills, S/A/B/C ratings, deploy anywhere

    HTML 1 1

  5. devspace devspace Public

    Forked from Waishnav/devspace

    Turn ChatGPT into Codex

    TypeScript 1

  6. claude-html-style-router claude-html-style-router Public

    Route any HTML deliverable to the right design system (10 genres) on top of a shared typographic foundation with first-class CJK correctness. Claude Code / Skills pack.

    Python 1 1